A layered software structure (platform) is a means of improving portability and maintainability of software in information processing and switching systems. The authors have proposed a CTRON benchmark program for eval...
详细信息
ISBN:
(纸本)0818645806
A layered software structure (platform) is a means of improving portability and maintainability of software in information processing and switching systems. The authors have proposed a CTRON benchmark program for evaluating the performance of various platforms. This paper examines the following aspects, based on the results of platform evaluations using the CTRON Benchmark program. (1) Methods for normalizing hardware performance (2) Performance differences depending on OS implementation. (3) Performance comparison between CTRON-specification systems and other realtime operating systems.
Recent evidence indicates that the exploitation of locality in dataflow programs could have a dramatic impact on performance. the current rend in the design of dataflow processors suggest a synthesis of traditional no...
详细信息
ISBN:
(纸本)0897915097
Recent evidence indicates that the exploitation of locality in dataflow programs could have a dramatic impact on performance. the current rend in the design of dataflow processors suggest a synthesis of traditional non-strict fine grain instruction execution and a strict coarse grain execution in order o exploit locality. While an increase in instruction granularity will favor the exploitation of locality within a single execution thread, the resulting grain size may increase latency among execution threads. In this paper, the resulting latency incurred through the partitioning of fine grain instructions into coarser grain threads is evaluated. We define the concept of a cluster of fine grain instructions to quantify coarse grain input and output latencies using a set of numeric benchmarks. The results offer compelling evidence that the inner loops of a significant number of numeric codes would benefit from coarse grain execution. Based on cluster execution times, more than 60% of the measured benchmarks favor a coarse grain execution. In 63% of the cases the input latency to the cluster is the same in coarse or fine grain execution modes. These results suggest that the effects of increased instruction granularity on latency is minimal for a high percentage of the measured codes, and in large part is offset by available intra-thread locality. Furthermore, simulation results indicate that strict or non-strict data structure access does not change the basic cluster characteristics.
In this paper we introduce the notion of task-delegation as compared to the delegation of rights only (rights-delegation). The reasoning behind the notion of task-delegation is presented, citing examples from the huma...
详细信息
ISBN:
(纸本)0818645806
In this paper we introduce the notion of task-delegation as compared to the delegation of rights only (rights-delegation). The reasoning behind the notion of task-delegation is presented, citing examples from the human world and from the realm of computers. This notion of task-delegation is captured in a 2-Phase Certification Server (2PCS) which we develop as a hybrid between Authentication Servers (based on shared key cryptosystems) and Certification Servers (based on public key cryptosystems). The 2PCS allows one node to task-delegate to another in a secure manner, and it achieves a high level of security due to its underlying strong cryptosystem. Within the framework of the TRON Architecture the features embodied within the 2PCS can be integrated within the CTRON architecture, either within a stand-alone server that mediates access to other CTRON servers (eg. file server) and other TRON components, or within existing servers to guard access to their corresponding resources. The first option represents a more manageable solution since cryptographic information (such as keys) need not be replicated. Other servers can then be treated as principals equivalent to the BTRON workstations. In either approach, we believe that the notion of task-delegation and the security features of the 2PCS can be a useful step towards integrating security into the TRON Architecture.
For porting of a C language program, compilers at the porting source and destinations must interpret the program identically. This can lead that a portability scope is defined by the set of compilers that will interpr...
详细信息
ISBN:
(纸本)0818645806
For porting of a C language program, compilers at the porting source and destinations must interpret the program identically. This can lead that a portability scope is defined by the set of compilers that will interpret a program identically. Similarly, a group of programs can be delineated whose interpretation will be the same within a set of compilers. This paper proposes a Porting Class which is a portability scope for the programs written in C language, and indicates the coding rules which must be observed to belong to each Porting Class. Methods are also indicated for checking whether a program adheres to those coding rules.
Many companies want to use modern statistical software improve the quality, cost and timeliness of their manufacturing processes and products. However, the best of tools is worthless if not used. This paper shows orga...
详细信息
Many companies want to use modern statistical software improve the quality, cost and timeliness of their manufacturing processes and products. However, the best of tools is worthless if not used. This paper shows organizations how to train their people simultaneously in both statistical process improvement techniques and statistical software tools, with minimum training time and costs. By presenting the techniques from the manufacturing perspective, and by using a training environment that is 100% hands-on and incorporates realistic beginning-to-end process scenarios, participants gain the confidence and skills to improve their processes and products with a minimum expenditure of their manufacturing resources. For those who design, teach, select, or participate in statistical training, the considerations presented here ensure the maximum benefit to the individuals and organization being trained.
Two strategies for supporting asynchronous interrupts are: the use of the processor's hardware interrupt system and the use of polling. The advantages of polling include: portability, simplicity, and low cost for ...
详细信息
ISBN:
(纸本)089791595X
Two strategies for supporting asynchronous interrupts are: the use of the processor's hardware interrupt system and the use of polling. The advantages of polling include: portability, simplicity, and low cost for handling interrupts. Unfortunately, polling has an overhead for the explicit interrupt checks inserted in the code. This paper describes balanced polling, a method for placing the interrupt checks which has a low overhead and also guarantees an upper bound on interrupt latency. This method has been used by Gambit (an optimizing native code compiler for Scheme) to support a number of features including multiprocessing and stack overflow detection. The overhead of balanced polling is less than for call-return polling which places interrupt checks at every procedure entry and exit. The overhead of call-return polling is typically 70% larger (but sometimes over 400% larger) than the overhead of balanced polling.
This paper presents knowledge-oriented programming, a framework for giving high-level knowledge-based descriptions of distributed protocols. We extend the notion of knowledge-based programs of [FHMV93] by allowing hig...
详细信息
ISBN:
(纸本)0897916131
This paper presents knowledge-oriented programming, a framework for giving high-level knowledge-based descriptions of distributed protocols. We extend the notion of knowledge-based programs of [FHMV93] by allowing high-level actions that are defined in terms of changing the state of knowledge of processes in the system. As a result, both the choice of which actions to perform and the actions themselves are defined explicitly in terms of the processes' knowledge. We concentrate on a class of high-level notify actions that can be used to abstract away details of communication. Examples are considered, including solutions to the atomic commitment problem and to the sequence transmission problem.
The implementation of sequential loops in dataflow computation had traditionally not received very much attention as it was assumed that most loops would be executed in parallel. This assumption was valid for earlier ...
详细信息
ISBN:
(纸本)089791595X
The implementation of sequential loops in dataflow computation had traditionally not received very much attention as it was assumed that most loops would be executed in parallel. This assumption was valid for earlier dataflow machines such as the MIT Tagged Token Dataflow Architecture (TTDA), Sigma-1 but not for the newest generation of dataflow machines including Monsoon, EM-4 and Epsilon-2. On the latter machines, sequential loops use less memory, and can execute in fewer instructions, albeit with lower parallelism than the parallel versions. This characterization of sequential and parallel loops suggests that programs should have parallel outer loops and sequential inner loops. The run time of sequential loops therefore become significant in the overall run time. We also found that previous implementations of sequential loops can incur fairly high overheads. In this paper, we present two new ways of implementing sequential loops that have lower overhead then previous methods. We studied this problem in the context of compiling Id for Monsoon.
We deal with conceptual problems associated with the methodology of application program system design and of unification of intelligent interaction with a user. Attention is mainly paid to substitution of the structur...
详细信息
We deal with conceptual problems associated with the methodology of application program system design and of unification of intelligent interaction with a user. Attention is mainly paid to substitution of the structural modular compositional programming method and to the user interface construction.
Some fundamental differences between Ada and C as the language for operating system description are listed up, specifically, those differences observed in translating the C-based MINIX operating system into Ada are co...
详细信息
Some fundamental differences between Ada and C as the language for operating system description are listed up, specifically, those differences observed in translating the C-based MINIX operating system into Ada are collected;there are also given concrete illustrative program examples.
暂无评论